Forward Deployed Engineer - Media Generation

Role overview This role owns the end-to-end benchmarking pipeline for leading image and video generation models, producing independent quality and performance comparisons for the AI industry. It is a deployed, process-driven engineering position that combines pipeline reliability work with external-facing technical communication to model providers.

Responsibilities - Generate image and video outputs across multiple models following standardized evaluation protocols - Design, launch, and manage human preference evaluation studies, including participant coordination and quality control - Process and analyze preference vote data to produce benchmark results - Serve as the primary technical contact for media generation model providers, answering methodology and result queries - Maintain reliable day-to-day operation of evaluation pipelines and data collection workflows - Communicate findings clearly to model providers, lab teams, and partner organizations

Requirements - Strong proficiency in Python for pipeline and data work - Detail orientation and a track record of running reliable, repeatable processes - Hands-on experience with at least one image or video generation system (e.g., image or video synthesis, diffusion models, or comparable media tooling) - Comfort working with external stakeholders as a technical point of contact - Background in data analysis, research operations, or a similar quantitative discipline

Nice to have - Familiarity with human evaluation methodologies or preference-based ranking systems - Experience in B2B SaaS, developer tools, or model evaluation platforms

Benefits and work setup - Full-time position based in San Francisco (on-site) or remote within the United States - Competitive compensation package that includes equity